Karina Nguyen is building Thoughtful, an AI research and product lab. She is also an angel investor and advisor to AI startups and neolabs, and backs projects in design, fashion, hospitality and the arts. Previously she was an AI researcher at OpenAI, working across post-training, practical alignment, reinforcement learning and product. There she led research for ChatGPT Canvas and contributed to the o-series reasoning models, GPT-4o, Tasks and SimpleQA. She joined Anthropic in summer 2022 and spent two years on post-training for Claude 1 through 3, research on model behavior such as honesty, harmlessness and reducing hallucinations, and early products including claude.ai and Claude in Slack. Earlier she worked in visual forensics and open-source intelligence, contributing to Pulitzer Prize-winning reporting, and interned in software engineering and product design at Primer.ai, Dropbox, Square, The New York Times and Fireflies.ai. She studied data science at UC Berkeley.